Diagnostic performance of the current risk-stratified approach with computed tomography for suspected choledocholithiasis and its options when negative finding

Background:Several studies evaluated the current guideline of the American Society for Gastrointestinal Endoscopy(ASGE)and reported only suboptimal accuracy.This study evaluated the diagnostic performance of the ASGE guideline based on computed tomography(CT)and role of endoscopic ultrasonography(EUS)and magnetic resonance cholangiopancreatography(MRCP)in patients with suspected choledocholithiasis but negative CT finding.Methods:Patients with suspected choledocholithiasis undergoing ERCP between January 2016 and January 2017 were retrospectively analyzed.All patients underwent CT to detect choledocholithiasis.EUS or MRCP was performed when the CT scan showed negative findings.Patients were classified into the high and intermediate-risk groups,based on predictors from the ASGE criteria.Results:Of 583 patients with suspected choledocholithiasis,340(58.3%)had stones on ERCP(65.9%in the high-risk group and 40.6%in the intermediate-risk group).The accuracy of ASGE guideline for CT was 63.98%(79.12%sensitivity,42.80%specificity)and 36.02%(20.88%sensitivity,57.20%specificity)in the high-risk and intermediate-risk groups,respectively.In 103 patients in the high-risk group underwent both CT and US,the accuracy of CT was higher than that of US for detecting choledocholithiasis(78.64%vs.53.40%),with a significant difference in area under the curve(AUC)(0.78 vs.0.59,P<0.001).Of 339 with negative CT finding,the accuracy of EUS was higher than that of MRCP(90.91%vs.82.76%),but with no significant difference in AUC(0.91 vs.0.83,P=0.347).Conclusions:CT-based ASGE guideline showed superior diagnostic performance than US for predicting choledocholithiasis.The diagnostic options,EUS or MRCP,with negative CT finding showed comparable performance.Therefore,the diagnostic modality should be selected based on availability,experience,cost,and contraindications.

Short Term Wind Power Prediction Using Wavelet Transform and ARIMA

A sustainable production of electricity is essential for low carbon green growth in South Korea. Although wind energy is unlimited in potential, both intermittency and volatility should be tackled for smart grid integration in future. To cope with this, many works have been done for wind speed and power forecasting. It is shown that statistical techniques are useful for short-term forecasting of wind power. This paper presents a statistical wind speed forecasting. The wavelet decomposition is employed as a de-noising technique. An illustration will be given by real-world dataset. According to the result, the MAD （mean absolute deviation） is improved as much as 10%.